A Region-Based Image Fusion Method Using the Expectation-Maximization Algorithm

We present a novel region-based image fusion method using a rigorous application of estimation theory. This method takes advantage of the similar intensity or texture in a region for fusion. A statistical image formation model using Gaussian mixture distortion is built for each region and the EM (expectation-maximization) algorithm is used in conjunction with the model to develop the region-level EM fusion algorithm to produce the fused image. Since in most applications of image fusion, objects carry the information of interest and regions can be used to represent objects, the region-based fusion approaches could be more meaningful than pixel-based methods. Our experiments demonstrate the efficiency of the proposed region-base fusion method and the advantages in dealing with region interface artifacts for concealed weapon detection and night vision applications.

[1]  Pramod K. Varshney,et al.  Enhancement and fusion of data for concealed weapon detection , 1997, Defense, Security, and Sensing.

[2]  E. Micheli-Tzanakou,et al.  Medical imaging fusion applications: An overview , 2001, Conference Record of Thirty-Fifth Asilomar Conference on Signals, Systems and Computers (Cat.No.01CH37256).

[3]  Edward H. Adelson,et al.  The Laplacian Pyramid as a Compact Image Code , 1983, IEEE Trans. Commun..

[4]  R.S. Blum,et al.  Multi-frame image fusion using the expectation-maximization algorithm , 2005, 2005 7th International Conference on Information Fusion.

[5]  Rick S. Blum,et al.  An Overview of Image Fusion , 2005 .

[6]  Ren C. Luo,et al.  Multisensor integration and fusion for intelligent machines and systems , 1995 .

[7]  Rick S. Blum,et al.  A statistical signal processing approach to image fusion for concealed weapon detection , 2002, Proceedings. International Conference on Image Processing.

[8]  Rick S. Blum,et al.  An adaptive spatial diversity receiver for non-Gaussian interference and noise , 1997, First IEEE Signal Processing Workshop on Signal Processing Advances in Wireless Communications.

[9]  Milan Sonka,et al.  Image Processing, Analysis and Machine Vision , 1993, Springer US.

[10]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[11]  Gemma Piella,et al.  A general framework for multiresolution image fusion: from pixels to regions , 2003, Inf. Fusion.

[12]  Lorenzo Bruzzone,et al.  Image fusion techniques for remote sensing applications , 2002, Inf. Fusion.

[13]  Gemma Piella,et al.  A region-based multiresolution image fusion algorithm , 2002, Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997).

[14]  Colin E. Reese,et al.  Comparison of additive image fusion vs. feature-level image fusion techniques for enhanced night driving , 2003, SPIE Optics + Photonics.

[15]  Alfred O. Hero,et al.  Space-alternating generalized expectation-maximization algorithm , 1994, IEEE Trans. Signal Process..

[16]  Jake K. Aggarwal,et al.  Multisensor Fusion for Computer Vision , 1993, NATO ASI Series.

[17]  R. Redner,et al.  Mixture densities, maximum likelihood, and the EM algorithm , 1984 .

[18]  Pramod K. Varshney,et al.  Concealed weapon detection: an image fusion approach , 1997, Defense + Security.

[19]  Peter J. Burt,et al.  Enhanced image capture through fusion , 1993, 1993 (4th) International Conference on Computer Vision.

[20]  Henk J. A. M. Heijmans,et al.  MULTIRESOLUTION IMAGE FUSION GUIDED BY A MULTIMODAL SEGMENTATION , 2002 .

[21]  Rick S. Blum,et al.  A categorization of multiscale-decomposition-based image fusion schemes with a performance study for a digital camera application , 1999, Proc. IEEE.